**** BEGIN LOGGING AT Wed Oct 01 02:59:59 2008 Oct 01 06:12:37 Hi all Oct 01 06:54:22 I am little familiar to OE. I want to build more than 1 independent packages { not all packages for which we can use "bitbake world"}. please suggest some option so that these packages could be built in single command. Oct 01 06:56:43 bitbake foo bar Oct 01 06:56:43 or if you have the same set of packages usually Oct 01 06:56:43 just create a task recipe Oct 01 06:56:48 that RDEPENDS on the packages you want Oct 01 07:00:15 good morning Oct 01 07:00:59 good morning Oct 01 07:01:44 Can anybody please tell where can I get the variable information in OE? Oct 01 07:04:10 While trying to build a simple package it will show the error message as failed with 256 Oct 01 07:04:55 testpackages.bb failed Oct 01 07:22:48 morning Oct 01 07:40:08 bonjour Oct 01 07:45:08 !oebug 3053 Oct 01 07:45:08 * * Bug 3053, Status: RESOLVED (WORKSFORME), Created: 2007-09-22 11:03 Oct 01 07:45:09 * * cesarb(AT)cesarb.net: ruby-1.8.5 fails to build with dash as /bin/sh Oct 01 07:45:10 * * http://bugs.openembedded.net/show_bug.cgi?id=3053 Oct 01 07:47:52 morning Oct 01 07:52:07 yo harald Oct 01 07:55:47 raster: ? Oct 01 07:56:07 morning Oct 01 07:56:08 :) Oct 01 07:56:29 raster: I am not Harald :D Oct 01 07:56:35 oooh damn Oct 01 07:56:41 u stole his nick! Oct 01 07:56:52 raster: really? Oct 01 07:56:55 yeah Oct 01 07:57:10 harald welte Oct 01 07:57:14 normally hrw Oct 01 07:57:15 :) Oct 01 07:57:27 01 09:56 -NickServ(NickServ@services.)- Information on Hrw (account Hrw): Oct 01 07:57:27 01 09:56 -NickServ(NickServ@services.)- Registered : Mar 15 10:59:47 2004 (4 years, 28 weeks, 4 days, 20:56:14 ago) Oct 01 07:58:33 so please do not make me laugh that hrw is Harald nickname Oct 01 08:01:32 ooh sorry no Oct 01 08:03:38 i could have sword harald uses hrw Oct 01 08:03:41 somewhere Oct 01 08:16:04 There are more dogs called Goofy. Oct 01 08:17:29 is there a magic beacons to paste some code on the wiki ? Oct 01 08:17:50 works on my corporate mediawiki but not here .. Oct 01 08:20:53 yop zecke ; i try to write some info about how to use lib bb Oct 01 08:21:24 ah cool Oct 01 08:21:39 i would like to make something usefull for the project Oct 01 08:22:26 :) Oct 01 08:22:27 aaah the mystery of the .bb Oct 01 08:23:03 no using bb library to make other dev , not only bbclasS ... Oct 01 08:23:30 oooh Oct 01 08:23:34 like bitbake itself Oct 01 08:23:36 not .bb files Oct 01 08:23:52 yeap Oct 01 08:27:01 http://wiki.openembedded.net/index.php/Bitbake Oct 01 08:27:25 i don't find a good way to paste code source on wiki Oct 01 08:39:14 morning all Oct 01 08:40:46 yop richard :) Oct 01 08:42:29 morning all, in org.openembedded.dev/packages/openchrome/xf86-video-openchrome_0.2.902.bb you can find a reference to file://configure-dri.patch, but i cannot find this patch anywhere Oct 01 08:42:33 can anyone help? Oct 01 08:48:21 Genesis:

code

maybe? Oct 01 08:48:36 exodos: I tried to build openchrome a few days ago too and discovered the same thing. Never got around to fixing the issue in my tree though. Oct 01 08:49:19 gg hrw :) Oct 01 08:49:34 * Genesis is a bit silly this morning Oct 01 08:55:46 !oebug 4508 Oct 01 08:55:48 * * Bug 4508, Status: NEW, Created: 2008-08-19 12:13 Oct 01 08:55:49 * * ronan(AT)aimao.org: new recipe for OpenCV - The Open Computer Vision Library Oct 01 08:55:49 * * http://bugs.openembedded.net/show_bug.cgi?id=4508 Oct 01 08:56:13 hail zecke Oct 01 09:04:24 right Oct 01 09:09:36 florian: good morning Oct 01 09:09:50 btw, I have a patch in poky's bitbake which stops recipes using SRCREV from making it into the cache. This may help the OM guys? Oct 01 09:10:20 hi pb__ Oct 01 09:10:49 03 07org.openembedded.dev * rb7f1f89d1e 10OE.dev/packages/linux/ (8 files in 3 dirs): linux-omap git: add MUSB patch to fix multiple bulk transfers, drop obsolete patches Oct 01 09:13:35 bonjour gregoiregentil :) Oct 01 09:17:18 03 07org.openembedded.dev * r310f2dce06 10OE.dev/packages/imagemagick/ (3 files in 2 dirs): imagemagick: add 6.4.4-1 Oct 01 09:35:03 what will happen when the final conversion to git kicks in? Oct 01 09:35:56 will the commit ids from the current test git.openembedded.org stay valid or will I have to rebase all local work? Oct 01 09:35:59 monotone to readonly Oct 01 09:44:25 Esben: rebase Oct 01 09:50:43 zecke: thanks Oct 01 09:54:31 hi Stecchino Oct 01 09:54:36 hi steliosk Oct 01 10:00:51 03 07org.openembedded.dev * re78c9a6d74 10OE.dev/packages/uclibc/ (uclibc-0.9.29/uClibc.distro uclibc_0.9.29.bb): uclibc: tweak debug opts Oct 01 10:00:51 03 07org.openembedded.dev * r0a52f9731d 10OE.dev/packages/mtd/mtd-utils_1.1.0.bb: mtd-utils: fix SRC_URI Oct 01 10:00:52 03 07org.openembedded.dev * r7dd5da46d9 10OE.dev/packages/git/git.inc: git: reorder vars Oct 01 10:00:54 03 07org.openembedded.dev * r07153c3313 10OE.dev/packages/gnash/gnash-minimal.inc: gnash minimal: make sound work Oct 01 10:00:55 03 07org.openembedded.dev * r7587f9a3c0 10OE.dev/packages/linux/linux_2.6.25.bb: linux 2.6.25: bump PR for avr32 defconfig changes Oct 01 10:00:58 03 07org.openembedded.dev * rd33e5f13b0 10OE.dev/classes/qmake2.bbclass: qmake2 bbclass: pass RCC var Oct 01 10:01:01 03 07org.openembedded.dev * rae4b42f253 10OE.dev/packages/samba/samba.inc: samba: make do_configure_prepend work with more recent versions Oct 01 10:01:04 03 07org.openembedded.dev * r661a353d78 10OE.dev/classes/packaged-staging.bbclass: packaged-staging bbclass: make relinking work Oct 01 10:04:48 hrw : hi ! Oct 01 10:12:53 * zecke sees a company confidential flash video that is linked from the front page of a website... *fail* Oct 01 10:13:45 hi all Oct 01 10:13:49 hi zecke Oct 01 10:39:11 zecke: heh Oct 01 10:56:33 03 07org.openembedded.dev * re652207a1d 10OE.dev/packages/perl/ (perl-native_5.8.8.bb perl.inc perl_5.8.8.bb): Oct 01 10:56:33 perl: fix references to STAGING_DIR/HOST_SYS, it needs to be STAGING_DIR_HOST for the upcoming MULTIMACH_TARGET_SYS changes Oct 01 10:56:33 * see http://lists.linuxtogo.org/pipermail/openembedded-devel/2008-July/005819.html Oct 01 11:40:47 What is the status of cross-linkage and staging-linkage in current .dev? Oct 01 11:41:03 are they supposed to be obsolete? Oct 01 11:42:17 perhaps current packaged staging have "broken" them? Oct 01 11:45:16 * * OE Bug 4590 has been created by nicola.mfb(AT)gmail.com Oct 01 11:45:18 * * ttf-liberation-0.2-r2 fails on do_install Oct 01 11:45:20 * * http://bugs.openembedded.net/show_bug.cgi?id=4590 Oct 01 11:47:17 * * OE Bug 4584 has been RESOLVED (FIXED) by rod(AT)whitby.id.au Oct 01 11:47:19 * * Unbreak ttf fonts Oct 01 11:47:21 * * http://bugs.openembedded.net/show_bug.cgi?id=4584 Oct 01 12:07:34 root Oct 01 12:07:49 Crofton|work: now type your password too! Oct 01 12:08:00 it won't take return Oct 01 12:08:17 * Crofton|work is logging in to the beagle Oct 01 12:08:36 1234 Oct 01 12:08:41 dooh ;-) Oct 01 12:09:06 hi steliosk Oct 01 12:09:59 there seems to be an issue with git tree updating. Bugs 4584 and 4590 weren't really issue, but only reflected the problem that monotone and git aren't synced. Oct 01 12:10:50 should I write to mailing list or does anyone here know about the git tree? Oct 01 12:11:23 apparently the master branch no longer updates Oct 01 12:11:35 http://gitweb.openembedded.net/?p=org.openembedded.dev.git;a=summary Oct 01 12:11:41 look at the heads near the bottom Oct 01 12:11:57 org.openembedded.dev not master Oct 01 12:12:01 not sure what is going on Oct 01 12:20:43 likewise : hi ! Oct 01 13:10:17 03 07org.openembedded.dev * r118c1db277 10OE.dev/classes/cross.bbclass: cross bbclass: sync with poky Oct 01 13:13:02 <[X]Spot> Hi Oct 01 13:13:07 <[X]Spot> http://stancho.com/log.do_compile.28701 Oct 01 13:14:40 <[X]Spot> Crofton|work sorry that I am disturbing you very often.. but I really don't know what's going on with this oe.. I hoped everything should be ok during the clean installation following the installation guide Oct 01 13:14:55 you are having really wierd erros Oct 01 13:15:10 what package is failing Oct 01 13:15:20 <[X]Spot> NOTE: package git-native-1.5.3-r2: task do_compile: started Oct 01 13:15:20 <[X]Spot> ERROR: function do_compile failed Oct 01 13:15:20 <[X]Spot> ERROR: see log in /OE/angstrom_stable//work/i686-linux/git-native-1.5.3-r2/temp/log.do_compile.28701 Oct 01 13:15:20 <[X]Spot> NOTE: Task failed: /OE/angstrom_stable//work/i686-linux/git-native-1.5.3-r2/temp/log.do_compile.28701 Oct 01 13:15:20 <[X]Spot> NOTE: package git-native-1.5.3-r2: task do_compile: failed Oct 01 13:15:20 <[X]Spot> ERROR: TaskFailed event exception, aborting Oct 01 13:15:22 <[X]Spot> NOTE: package git-native-1.5.3: failed Oct 01 13:15:24 <[X]Spot> ERROR: Build of /OE/org.openembedded.dev/packages/git/git-native_1.5.3.bb do_compile failed Oct 01 13:15:26 <[X]Spot> ERROR: Task 799 (/OE/org.openembedded.dev/packages/git/git-native_1.5.3.bb, do_compile) failed Oct 01 13:15:28 <[X]Spot> NOTE: Waiting for 1 active tasks to finish Oct 01 13:15:30 <[X]Spot> NOTE: 1: /OE/org.openembedded.dev/packages/gcc/gcc-cross_4.3.1.bb, do_populate_staging (28700) Oct 01 13:15:32 <[X]Spot> NOTE: package gcc-cross-4.3.1-r15: task do_populate_staging: completed Oct 01 13:15:34 it really looks like you have screwed up something very fundamental Oct 01 13:15:34 <[X]Spot> NOTE: package gcc-cross-4.3.1: completed Oct 01 13:15:36 <[X]Spot> NOTE: Tasks Summary: Attempted 284 tasks of which 284 didn't need to be rerun and 1 failed. Oct 01 13:15:38 <[X]Spot> ................. Oct 01 13:15:40 <[X]Spot> git-native Oct 01 13:15:51 are you build beagle from the stable branch or something like that Oct 01 13:15:57 not using angstrom-2208 Oct 01 13:16:04 ~pastebin Oct 01 13:16:04 [~pastebin] A "pastebin" is a web-based service where you can paste anything over 3 lines without flooding the channel. Here are links to a few : http://www.pastebin.com , http://pastebin.ca , http://channels.debian.net/paste , http://paste.lisp.org , http://www.rafb.net/paste Oct 01 13:16:40 03 07org.openembedded.dev * r80119eac7f 10OE.dev/packages/gcc/ (gcc-common.inc gcc-package-sdk.inc gcc-package-target.inc): gcc: sync with poky Oct 01 13:17:30 <[X]Spot> I just followed http://www.angstrom-distribution.org/building-%C3%A5ngstr%C3%B6m ... but instead of doing mtn.. I take the last dev : git clone git://git.openembedded.net/org.openembedded.dev.git Oct 01 13:18:09 can you pastebin your local.conf Oct 01 13:20:33 <[X]Spot> Here it is my /OE/build/conf/local.conf Oct 01 13:20:37 <[X]Spot> http://stancho.com/local.conf Oct 01 13:22:53 [X]Spot, use a pastbin ,,, Oct 01 13:23:17 [X]Spot, just for grain can you try for beagelbaord Oct 01 13:31:03 <[X]Spot> ok Oct 01 13:31:09 <[X]Spot> sorry Oct 01 13:33:35 <[X]Spot> maybe I should remove /OE/angstrom_stable/ after changing the machine type ? Oct 01 13:38:12 [X]Spot, good idea Oct 01 13:38:21 that is your tmpdir, right? Oct 01 13:43:56 good morning Oct 01 13:46:16 Is there someone using firefox 3.0.1? Oct 01 13:46:38 <[X]Spot> Crofton|work that's right Oct 01 13:46:52 <[X]Spot> I am now installing pastebin ;) Oct 01 13:47:00 Mine is lacking the icons for arrows and like Oct 01 13:47:08 ~pastebin Oct 01 13:47:09 [~pastebin] A "pastebin" is a web-based service where you can paste anything over 3 lines without flooding the channel. Here are links to a few : http://www.pastebin.com , http://pastebin.ca , http://channels.debian.net/paste , http://paste.lisp.org , http://www.rafb.net/paste Oct 01 13:47:13 it is a web page Oct 01 13:47:17 I'm using today's tip repository Oct 01 13:47:18 just use an exsiting one Oct 01 13:49:21 * otavio mailed ml Oct 01 13:54:10 Crofton|work " hi ! Oct 01 13:54:38 steliosk, hello Oct 01 13:54:45 I didn't make it to Phoenix Oct 01 13:54:54 still rumours I may go Oct 01 13:55:32 Crofton|work : Was an interesting conference... from all aspects :) Oct 01 13:55:41 great Oct 01 13:55:47 got to see my fist ice hokey match :) Oct 01 13:55:53 heh Oct 01 13:56:04 with the outside temp being 100F? Oct 01 13:56:07 ice hokey in the middle of the Arizona desert.... Oct 01 13:56:16 yeah something like that Oct 01 13:57:33 across the hotel there was a beach volley tournament, with the US Olympic team and lots of other nice stuff ;) Oct 01 13:57:54 sounds like an excellent trip :) Oct 01 13:58:22 these were the extra-conference activities :) Oct 01 13:58:43 <[X]Spot> Crofton|work I was installing it on my server ;) Oct 01 13:59:18 yeah it was fun.... more than usual for these kind of conferences, to tell you the truth, but the company was good. Oct 01 13:59:28 guys, to enable locale support, arm/uclibc/oe.stable, is it enough to set USE_NLS to "yes" ? Oct 01 13:59:37 or is some other magic needed in addition? Oct 01 14:00:55 Crofton|work : Keynotes where interesting.... Asterisk will support Skype Oct 01 14:02:47 <[X]Spot> Crofton|work http://stancho.com/pastebin/ Oct 01 14:02:47 <[X]Spot> :) Oct 01 14:04:03 <[X]Spot> ex: http://stancho.com/pastebin/1 Oct 01 14:04:55 We need the context for this error Oct 01 14:05:13 remember, I am running on F8 and not having any issues Oct 01 14:05:32 03 07org.openembedded.dev * r63b1dcdabc 10OE.dev/packages/libsdl/libsdl-native_1.2.11.bb: libsdl-native: disable parallel make Oct 01 14:05:52 if this is git native, naybe installing curl-devel on the host would help Oct 01 14:06:27 cleaning disk from old builds is argh... Oct 01 14:06:48 I found ~half year old builds :) Oct 01 14:07:57 hrw, do you need the disk space for pictures? Oct 01 14:08:33 Crofton|work: no - I just had less then 100GB free on harddrive Oct 01 14:09:33 I got ~160GB back now Oct 01 14:10:40 an another old build... this time just 700MB Oct 01 14:13:56 <[X]Spot> Crofton|work I am now rebuilding it with MACHINE="beagleboard" Oct 01 14:14:35 it is git-native that isi the problem? Oct 01 14:15:22 it does have a dependency on curl-native, which should provide the headers Oct 01 14:15:53 and under fedora9 it can fail if perl-ExtUtils-MakeMaker is not installed (poky has fix) Oct 01 14:20:40 <[X]Spot> Crofton|work that's right Oct 01 14:40:03 can anyone tell me what is the "OE way" of enabling UCLIBC_HAS_LOCALE in uclibc? I keep running into problems :P Oct 01 14:40:33 Jin^eLD: it should be enabled by default doesnt it Oct 01 14:41:33 khem: I think it depends (or should?) on the USE_NLS variable Oct 01 14:41:55 but if I set it to "yes" and then build, and look at .config in work/xxx/uclibc - then it is not set there Oct 01 14:42:05 this is on stable, I am retrying on dev to see if there is any difference Oct 01 14:43:02 oh yes it needs USE_NLS set Oct 01 14:43:22 if I set it manually, i.e. force the setting either via uClibc.distro, then uclibc does not compile Oct 01 14:43:33 ugh Oct 01 14:43:36 we might fix it Oct 01 14:43:43 whats the problem Oct 01 14:43:47 you are seeing Oct 01 14:43:57 well, I did set USE_NLS to "yes" but as I mentioned - the .config setting still shows it as not set Oct 01 14:44:29 let me quickly retry with .dev first, problem is that uclibc does not find some headers (which are there) Oct 01 14:44:37 let me see if I can find a paste Oct 01 14:46:37 ${@["s,.*UCLIBC_HAS_LOCALE.*,# UCLIBC_HAS_LOCALE is not set,;", ""][bb.data.getVar("USE_NLS", d, 1) == "yes"]} Oct 01 14:46:40 should set it Oct 01 14:46:43 http://pastebin.mozilla.org/545671 Oct 01 14:46:55 this is what I see if I "force" UCLIBC_HAS_LOCALE manually Oct 01 14:47:10 yes I know, it *should* but that does not seem to work out for some reason Oct 01 14:47:39 that's how I found out I was supposed to set USE_NLS in the first place Oct 01 14:48:04 I wonder why it does not work though.. but that's a different problem... because once it is enabled things go wroing when building Oct 01 14:48:26 if you look at the log.. this header: .//include/bits/uClibc_ctype.h - it is there Oct 01 14:51:32 hmmm Oct 01 14:52:05 in your build tree right ? Oct 01 14:52:07 my .dev is still building, I'm curious to see if USE_NLS produced the correct setting there Oct 01 14:52:50 yes, in build/angstorm/work/uclibc-blah-r8/uclibc-0.9.29/ Oct 01 14:53:36 http://blog.haerwu.biz/2008/10/01/how-much-ramhdd-is-enough/ Oct 01 14:54:07 OK, .dev is now building uclibc, I checked worked driectory - .config has # UCLIBC_HAS_LOCALE is not set Oct 01 14:54:23 and I do have USE_NLS = "yes" in my local.conf Oct 01 14:54:32 so that'd be problem #1 Oct 01 14:55:56 Jin^eLD: Try to use USE_NLS=1 Oct 01 14:56:13 ok, let me try that.. Oct 01 14:57:16 this header is not found thats the problem all these defines are in uClibc_config.h Oct 01 14:58:05 hm? but the header was there, it was some symlink but it was there.. Oct 01 14:58:20 I'm rebuilding with USE_NLS=1 now, let's see Oct 01 14:58:38 I dont understand why the header is not being picked Oct 01 14:58:47 me neither Oct 01 14:59:18 I tried to fix it by adding -I$(top_builddir) to the CFLAGS and that worked, but then another header was not found Oct 01 14:59:37 so something must have gone wrong on a different level.. something got screwed no idea where Oct 01 14:59:56 and btw this time I build for arm... so I wonder how noone else hit this problem Oct 01 15:01:56 hmmm I think top_builddir is busted in your case Oct 01 15:04:32 top_builddir=../../ Oct 01 15:04:41 thats what it should be in that place Oct 01 15:04:58 why would the UCLIBC_HAS_LOCALE setting cause such a havoc? Oct 01 15:06:36 I am *guessing* you are hitting a build block in uclibc which is not known yet Oct 01 15:06:53 bye Oct 01 15:07:13 hi mickey|tw Oct 01 15:07:14 khem: unchartered territory? :) Oct 01 15:07:22 hehe yes Oct 01 15:07:49 and I am the "lucky" one because noone else bothered to check if USE_NLS actually made a difference? Oct 01 15:07:55 hey pb__ Oct 01 15:08:37 khem: USE_NLS=1 - same stuff, # UCLIBC_HAS_LOCALE is not set Oct 01 15:09:42 http://wiki.openembedded.net/index.php/Bitbake Oct 01 15:10:03 now i try to use persist_data :) Oct 01 15:10:37 feel free to put the page on developer section in wiki homepage Oct 01 15:15:24 khem: well.. how would I debug the USE_NLS thing? Oct 01 15:15:42 does anybody know a good tool to do full-text indexing of maildir folders? I am running out of patience with the internal search in evolution, which is teh suck over imap. Oct 01 15:16:15 pb__: mairix? Oct 01 15:19:49 thanks, that sounds just like what I need Oct 01 15:22:42 Jin^eLD: Which distro are you using Oct 01 15:22:52 fedora 6 and 7 Oct 01 15:23:01 Jin^eLD: I mean OE distro :) Oct 01 15:23:04 ooh Oct 01 15:23:20 angstrom-2008.1 in .dev Oct 01 15:23:32 ok then do this ./distro/include/angstrom-uclibc.inc Oct 01 15:23:39 and angstrom-2007.12 in stable Oct 01 15:23:57 change USE_NLS ?= "no" Oct 01 15:24:02 to :USE_NLS ?= "yes" Oct 01 15:24:28 ok let me try.. Oct 01 15:24:42 or may be add USE_NLS_uclibc ?= "yes" Oct 01 15:24:45 just to test Oct 01 15:25:14 btw can I check with bitbake -i and getvar after doing the change or should I rebuild again? Oct 01 15:27:10 ...rebuilding... Oct 01 15:28:02 Jin^eLD: just rebuilding uclibc should be sufficient Oct 01 15:28:25 ok.. Oct 01 15:28:53 well too late, I wiped the build dir before your message :) but should not take long, its a fast machine Oct 01 15:39:13 quick question... my root filesystem is a jffs2 image and I'm running into a problem at boot where I run out of space on my root device Oct 01 15:39:21 how do I control the size of the image? Oct 01 15:39:48 or is there a different approach I should consider... Oct 01 15:40:16 khem: same result, # UCLIBC_HAS_LOCALE is not set Oct 01 15:41:48 03 07org.openembedded.dev * rc6dc897614 10OE.dev/packages/cacao/ (3 files in 2 dirs): cacao 0.99.3: Added two patches to fix problems with ARM targets. Oct 01 15:42:54 macdonalder: er, install fewer packages into the image? there isn't really any way to reduce the image size other than shipping fewer/smaller files or increasing the compression ratio. Oct 01 15:44:19 is it possible to increase the size of the image though? Oct 01 15:44:46 macdonalder: I believe the physical size of your flash will be the limit? Oct 01 15:45:06 or is it automatically configured to be the exact size of the contents of the filesystem? Oct 01 15:45:27 the output .jffs2 file will be the exact size of the contents of the filesystem. Oct 01 15:45:45 you can add empty space to the end if you want but there's no point: any blank area in your physical flash will automatically be used as free space anyway Oct 01 15:46:18 if you find that you don't have any space free at boot time then either you need to get bigger flash chips, or you need to reduce the size of the data in your image. Oct 01 15:46:23 ahhh... then perhaps I'm doing something wrong then Oct 01 15:47:10 khem: tried both, USE_NLS=1 and also "yes", getvar returns it correctly Oct 01 15:47:17 I'm using liveramdisk-image which uses that initramfs-jffs2 package Oct 01 15:47:19 khem: but the setting is just not set in .config Oct 01 15:47:48 its init script just cats the jffs2 to an mtd block device Oct 01 15:48:11 oh, right, then you need to make the mtd block device (or, rather, the ramfs that backs it) bigger Oct 01 15:48:23 ahhh ok Oct 01 15:49:15 I'll hunt around for how to do that then... if I still have problems I may be back Oct 01 15:49:17 thanks ;) Oct 01 15:52:53 khem: I don't know perl but I can only assume that this configmangle thing simply does not work correctly Oct 01 16:15:32 how does this ${@[bb.data.getVar ... stuff translate to shell variables? Oct 01 16:16:26 do_configure is a shell task, right? Oct 01 16:16:43 how would I access OE variables there? like the USE_NLS one? Oct 01 16:18:33 03 07org.openembedded.dev * r21b543e3ca 10OE.dev/packages/dbus/dbus-1.0.3/sanitize-reply-timeout.patch: dbus 1.0.3 add sanitize-reply-timeout patch also in its directory Oct 01 16:23:25 hm I think I figured it out Oct 01 16:24:19 <[X]Spot> Crofton|work Oct 01 16:24:25 <[X]Spot> the issue with beagleboard machine type Oct 01 16:24:28 <[X]Spot> with git-native Oct 01 16:24:29 <[X]Spot> http://stancho.com/pastebin/2 Oct 01 16:45:01 khem: I made sure the .config is set properly now (on idea why this sedding via perl did not work) Oct 01 16:45:08 khem: and now I of course hit the compile problem... Oct 01 16:50:18 * * OE Bug 4591 has been created by linuxrulez(AT)land.ru Oct 01 16:50:20 * * NOTE: package linux-handhelds-2.6-2.6.21-hh20-r20: task do_compile failed Oct 01 16:50:22 * * http://bugs.openembedded.net/show_bug.cgi?id=4591 Oct 01 17:02:29 Hello Oct 01 17:03:18 Is there anybody who supports magician platform in oe? Oct 01 17:12:18 I want to be a tester for magician platform and do what I can to help for building/testing/patching Oct 01 17:13:41 lastik_, maybe try emailing the list Oct 01 17:13:54 which one? Oct 01 17:14:22 Openembedded-devel@lists.openembedded.org ? Oct 01 17:17:23 yeah Oct 01 17:17:36 that may be a better way to find people working on the magiciain Oct 01 17:19:12 \msg Crofton|work ok, thanks! Oct 01 17:28:05 03 07org.openembedded.dev * rb78e02f465 10OE.dev/: Oct 01 17:28:05 merge of '7e947a0c90014aa35b2d54a97a35dac21ecd4545' Oct 01 17:28:05 and 'ede2922a2a1c1c3eb84491c1cb019edc21f2729e' Oct 01 17:28:06 03 07org.openembedded.dev * r8e90d181b9 10OE.dev/packages/sharp-binary-only/sharp-flash-header-c700.bb: sharp-flash-header: make machine specific Oct 01 17:28:06 03 07org.openembedded.dev * r2ffc61952d 10OE.dev/packages/gnome/gnome-games_2.24.0.bb: gnome-games_2.24.0.bb : Disable aisleriot until it really builds. Oct 01 17:49:12 Is there a way I can force OE to build an image of unstripped executables? Oct 01 17:53:01 INHIBIT_DEFAULT_STRIP=1 in local.conf? Oct 01 17:53:04 just a guess Oct 01 17:56:08 looks like DEBUG_BUILD and INHERIT_PACKAGE_STRIP Oct 01 17:57:18 * * OE Bug 4592 has been created by jin(AT)mediatomb.cc Oct 01 17:57:21 * * USE_NLS variable has no effect for uClibc configuration when building on Fedora 6/7 host Oct 01 17:57:23 * * http://bugs.openembedded.net/show_bug.cgi?id=4592 Oct 01 17:59:19 * * OE Bug 4593 has been created by jin(AT)mediatomb.cc Oct 01 17:59:21 * * building uClibc fails when USE_NLS is set to "yes" Oct 01 17:59:22 * * http://bugs.openembedded.net/show_bug.cgi?id=4593 Oct 01 18:05:01 INHIBIT, not INHERIT, but yes. Oct 01 18:05:52 :) Oct 01 18:06:03 I type poorly Oct 01 18:06:18 I need a global oprofile Oct 01 18:33:49 what does openembedded use in replacement of "xset" ? Oct 01 18:34:13 to change the auto-repeat rate on the keyboard Oct 01 18:35:44 03 07org.openembedded.dev * r7da5226fa6 10OE.dev/: Oct 01 18:35:44 merge of '4f2b8fbe38be5a535656d7d00413d303f9a4975d' Oct 01 18:35:44 and 'd3b1203776970588892a46acdd9df1e18c0c5d37' Oct 01 18:35:46 03 07org.openembedded.dev * re56c661a3c 10OE.dev/packages/gnome/gnome-games_2.24.0.bb: disapproval of revision '7e947a0c90014aa35b2d54a97a35dac21ecd4545' Oct 01 18:40:02 echelon: that would be a question for your distro. oe itself just supplies a bunch of packages, it doesn't choose one over another. Oct 01 18:40:19 oh :\ Oct 01 18:40:48 pb_: will which utils does oe supply? Oct 01 18:40:51 *well Oct 01 18:40:58 other than xset Oct 01 18:41:22 I don't know of any others, but that's not to say there aren't any. Oct 01 18:41:53 personally I just use xset. Oct 01 18:42:13 echelon: xset :) Oct 01 18:42:13 i don't have xset :\ Oct 01 18:42:34 well, like I say, that would be an issue that you should take up with your distro people. Oct 01 18:42:36 khem: Hello Oct 01 18:43:07 khem: you said on ml that firefox works great over there; good. I'd like to know if you have tested it in x86 based target? Oct 01 18:43:23 my distro is as good as dead, there's no one i can contact :\ Oct 01 18:43:23 khem: mine target is i586 and it doesn't shows the icons Oct 01 18:43:46 echelon: what distro are you using? Oct 01 18:44:23 pb_: zipit z2 Oct 01 18:44:39 ah. I've never even heard of that one. Oct 01 18:44:39 heh Oct 01 18:44:57 linux.zipitwireless.com Oct 01 18:45:20 of course, if you want to build xset for yourself, the files are right there in oe. Oct 01 18:45:37 otherwise, if your distro isn't maintained anymore, it sounds like you might just be SOL. Oct 01 18:45:47 i already have one i just can't remember what it's called Oct 01 18:46:04 ah, I see Oct 01 18:46:19 * * OE Bug 4594 has been created by linuxrulez(AT)land.ru Oct 01 18:46:21 * * qte-mt-static-2.3.10-r8 do_configure failed Oct 01 18:46:23 * * http://bugs.openembedded.net/show_bug.cgi?id=4594 Oct 01 18:46:31 you could try asking the x dudes in #freedesktop, they might be able to help you guess Oct 01 18:47:29 hrm, xset is packaged with X server? Oct 01 18:48:45 ah, that's why Oct 01 18:48:50 i don't have X :P Oct 01 18:49:10 so there's not auto-repeat setting for just the terminal? Oct 01 18:49:20 s/not/no Oct 01 18:51:38 not that I know of Oct 01 18:55:09 jo Oct 01 19:00:13 alright, thanks Oct 01 19:05:55 yop :) Oct 01 19:06:08 hi genesis Oct 01 19:08:18 pb_: kbdrate :) Oct 01 19:08:29 aha Oct 01 19:08:31 very good Oct 01 19:10:38 RP, : persist_data base is only used by the fetcher module yet , isn't it ? Oct 01 19:10:44 actually, kbd_mode Oct 01 19:10:59 kbd_mode was patched to include kbdrate in it Oct 01 19:11:59 hi pb Oct 01 19:14:03 what we need is a way to clean everything for machine and target, but not tool chain and native :) Oct 01 19:17:17 Crofton|work, i tried removing things from [work|staging|deploy|stamps]/[host|target] and bitbake freaked out. Oct 01 19:17:30 i had to rebuild everything from scratch Oct 01 19:17:42 kgilmer, I'm not sure you could do it in a useful way :) Oct 01 19:17:47 quad core is the answer Oct 01 19:18:14 I'm running a build on one core of this machine and it takes forever Oct 01 19:18:24 mainly to try and find obscure depends issues Oct 01 19:19:03 yeah... are you talking about the speed of the whole build or just bitbake loading itself? Oct 01 19:19:49 the world Oct 01 19:20:01 beagleboard-demo-image is very bloated Oct 01 19:20:08 but makes a good stress test Oct 01 19:21:18 * mwester found a lot of packages that wouldn't build reliably with parallel builds when attempting 4 cores :) Oct 01 19:21:34 mwester, I think there are missing depends Oct 01 19:22:11 Ah - you mean multiple bitbake tasks at once, then. Oct 01 19:22:24 since bb demo image has many packages, there are more ways to schedule something before somethign it depends on gets buit; Oct 01 19:22:25 I tried that and had trouble with the fetchers, and had to disable that. Oct 01 19:22:28 right Oct 01 19:27:27 re Oct 01 19:27:55 re marcin Oct 01 19:43:19 * * OE Bug 4595 has been created by mad_soft(AT)inbox.ru Oct 01 19:43:21 * * new recipe - evtest-1.23 Oct 01 19:43:23 * * http://bugs.openembedded.net/show_bug.cgi?id=4595 Oct 01 20:03:45 hmm.. Oct 01 20:03:55 is angstrom distro now unmaintained? Oct 01 20:04:07 * Jay7 looking at mtn update log Oct 01 20:04:34 ah Oct 01 20:04:40